7 Comments
User's avatar
Jack's avatar
Mar 22Edited

The problem with history is, can we really define "explanatory power" when N=1? People weave compelling stories about why lead plumbing caused the downfall of Rome, etc. but it always feels just-so to me. I wonder if it says more about the biases of the author (and their audience) than it does about Romans.

The randomness of LLM results may just be its peculiar way of saying, "I don't know." RLHF strongly penalizes the models for not giving answers.

Dave92F1's avatar

Would be easier to parse if LLM scores and polls were on normalized scales.

Phil Getts's avatar

I think 1700-1900 is too big a slice of time. At least divide it into Enlightenment and Industrial Revolution. 1789 would make a tolerable dividing line. I would more generally take some degrees of freedom out of the number of causes, and put them into finer time resolution.

Phil Getts's avatar

I've forgotten what "I asked a set of polls" means. Are you buying online polls? How do you do that?

Gareth's avatar

It would be interesting to know what prompt you used, and whether you (implicitly or explicitly) gave guidance as to which geographical, socioeconomic, or ethnic group to focus on. Also I wonder whether the results would be the different if the question was asked in French or Arabic, for example. I don’t know if this says more about the biases in the question or in the models.

Robin Hanson's avatar

I shared the LLM sessions in links in the post.